Artificial Synapses can classify image. This is a great development in the continuous quest towards artificial intelligence. There are 100 artificial synapses that are created by the researchers in UC Santa Barbara’s Department of Electrical and Computer Engineering and this synapsis can complete uncomplicated human task as it can classify image.
This development may be small compared on how great the remaining works mean this is what the professor of electrical and computer engineering, Dmitri Strukov said in a statement, “It’s a small, but important step.”
Precisely, as development progress it will be a matter of time that the circuitry may eventually be scaled to rival a human’s brain which has upwards of one quadrillion synaptic connections, according to the researchers.
The brain can accomplish certain functions — such as classifying the shapes of letters, and their relative positions to each other to derive meaning as you read this — in a fraction of a second. It would take a computer far more time and energy to perform these same tasks, further explained by the researchers.
Letters Z, V, N and their names can be classified by the circuit with a neural network, even if the images were stylized in different ways plus saturated noise still the circuit accurately recognized it as shown in the demonstration by researchers.
In addition a statement has been made the researcher Merrikh-Bayat, “While the circuit was very small compared to practical networks, it is big enough to prove the concept of practicality, “and he added that “As more solutions to the technological challenges are proposed the technology will be able to make it to the market sooner.”
A certain terminology called memristor is the key to this technology. Memristor is memory plus resistor combined the resistance of this electronic component changes when the direction of electrical charge flow changes. This means that memristors work based on iconic movement which is similar to the way human neural cells generate electronic signals.
With regards to memristors lead researcher Mirko Prezioso stated that “”Classical computers will always find an ineluctable limit to efficient brain-like computation in their very architecture,” he added “This memristor-based technology relies on a completely different way inspired by biological brain to carry on computation.”
According to researchers this technology is significant as it can be potentially used to improve fields like the medical field for instance, on the other hand, however the human brain is so complex that it will take vast of memristors supply so it can become like human brain.